SPIN Unprocessed July 9, 2026 ai_technology research
At-Grok Is Not Converged:A Measurement-Validity Audit for Grokking Representation Metrics
View original on arxiv.orgOverview
arXiv:2607.06639v1 Announce Type: new Abstract: On modular arithmetic, a network's embedding keeps compressing for tens of thousands of steps after it has already generalized. Reading effective rank at the grokking transition overstates the converged value by 3-5x on an MLP, and by 1.3-1.5x on a transformer trained to convergence; on the MLP it also erases which cells compress at all. Compression lags the accuracy transition by an amount on the order of the time-to-grok, at least 10,000 steps, r
SpinGraph analysis pending — check back after processing.
Ask AI about this story
Opens with the SpinGraph .md URL and structured context — one click, prompt included.
More from arXiv Machine Learning
View all →- UASPL: Uncertainty-Aware Self-Paced Learning with Evidential Neural Networks
- Optimized Instance Alteration for Explaining and Assessing Robustness of Classifiers
- When Do Geometric Algebra Layers Beat Scalarization? A Controlled Study on SO(3)-Equivariant Vector Laws
- Does Demand Response Increase Vulnerability to Cyber Attacks by Adversarial Data Modifications?
- When Certificates Fail: A Unified Safety Framework for Embedded Neural Interface Models
- STST-JEPA: Shallow-Target Spatio-Temporal Joint Embedding Prediction Architecture For EEG Self-Supervised Learning
Markdown (.md) · JSON-LD schema (.json) · Machine-readable for AI & GEO